Graph Neural Network vs Feature-based Folk Music Evolution Analysis

Résumé En

This paper explores changes in a novel graph-structured corpus of British folk music across time, to discover whether any evidence for evolution can be found. Feature-based approaches are compared to graph-neural network (GNN) models. Firstly, a large dataset of over 13,000 dated British folk tunes is collected and pitch and rhythm vectors are extracted. This dataset is made publicly available for future research 1 . 1000 tunes are considered henceforth in two datasets of even class distribution, with tunes grouped into 50-year or 25-year time periods respectively. Exploratory analysis is undertaken with the K-means algorithm on pitch and rhythm musicological descriptors extracted from each tune, revealing ill-defined clusters with significant overlap, implying that any differences across time periods are more high-dimensional than can be represented by simple features. However, clusters do seem to indicate that there are broad differences across 100-year periods. A graph based upon similarity between tunes is then constructed by creating edges using Euclidean distance between tune-vectors. Louvain community detection illustrates ill-defined communities in terms of time-period, with no clear evolutionary trends. Graph Convolutional Neural Network (GCN) and GraphSAGE models are trained on the two datasets and are found to perform above chance for detecting the time period of tunes. Peak performance reaches 57% accuracy for the GCN model trained on the 50-year class dataset, indicating differences between time periods within British folk music. However, evidence for evolution is tenuous. The GCN embedding space approximately indicates that classes between 1700 and 1900 are chronologically ordered, yet we do not see consistent misclassification to nearby time periods.